ConsistTL: Modeling Consistency in Transfer Learning for Low-Resource Neural Machine Translation

TL;DR

ConsistTL introduces a novel transfer learning approach for low-resource neural machine translation that continuously guides the child model during training by enforcing prediction consistency with a parent model, leading to significant performance gains.

Contribution

It proposes a dynamic transfer learning method, ConsistTL, that maintains ongoing knowledge transfer during training, unlike static prior methods.

Findings

Achieves up to 1.7 BLEU improvement on WMT17 Turkish-English benchmark.

Enhances inference calibration of the child model.

Demonstrates effectiveness across five low-resource NMT tasks.

Abstract

Transfer learning is a simple and powerful method that can be used to boost model performance of low-resource neural machine translation (NMT). Existing transfer learning methods for NMT are static, which simply transfer knowledge from a parent model to a child model once via parameter initialization. In this paper, we propose a novel transfer learning method for NMT, namely ConsistTL, which can continuously transfer knowledge from the parent model during the training of the child model. Specifically, for each training instance of the child model, ConsistTL constructs the semantically-equivalent instance for the parent model and encourages prediction consistency between the parent and child for this instance, which is equivalent to the child model learning each instance under the guidance of the parent model. Experimental results on five low-resource NMT tasks demonstrate that ConsistTL results in significant improvements over strong transfer learning baselines, with a gain up to 1.7 BLEU over the existing back-translation model on the widely-used WMT17 Turkish-English benchmark. Further analysis reveals that ConsistTL can improve the inference calibration of the child model. Code and scripts are freely available at https://github.com/NLP2CT/ConsistTL.